keras-team/keras · error · ValueError

`validation_split` must be strictly between 0 and 1. Receiv

Error message

`validation_split` must be strictly between 0 and 1.  Received: {validation_split}

What it means

validation_split, when truthy, must be strictly between 0 and 1 (exclusive). 0, 1, negative values, or values >= 1 raise this error.

Source

Thrown at keras/src/legacy/preprocessing/image.py:1021

        if data_format not in {"channels_last", "channels_first"}:
            raise ValueError(
                '`data_format` should be `"channels_last"` '
                "(channel after row and column) or "
                '`"channels_first"` (channel before row and column). '
                f"Received: {data_format}"
            )
        self.data_format = data_format
        if data_format == "channels_first":
            self.channel_axis = 1
            self.row_axis = 2
            self.col_axis = 3
        if data_format == "channels_last":
            self.channel_axis = 3
            self.row_axis = 1
            self.col_axis = 2
        if validation_split and not 0 < validation_split < 1:
            raise ValueError(
                "`validation_split` must be strictly between 0 and 1. "
                f" Received: {validation_split}"
            )
        self._validation_split = validation_split

        self.mean = None
        self.std = None
        self.zca_whitening_matrix = None

        if isinstance(zoom_range, (float, int)):
            self.zoom_range = [1 - zoom_range, 1 + zoom_range]
        elif len(zoom_range) == 2 and all(
            isinstance(val, (float, int)) for val in zoom_range
        ):
            self.zoom_range = [zoom_range[0], zoom_range[1]]
        else:
            raise ValueError(
                "`zoom_range` should be a float or "

View on GitHub (pinned to 7a34a03db6)

Solutions

1. Pass a fraction like 0.2 (0 < v < 1)
2. Omit validation_split entirely if no split is needed
3. If percentage in config, divide by 100 before constructing

Example fix

// before
ImageDataGenerator(validation_split=20)
// after
ImageDataGenerator(validation_split=0.2)

assert validation_split is None or 0 < validation_split < 1, validation_split

def valid_split(v): return v is None or (isinstance(v,(int,float)) and 0 < v < 1)

try: ImageDataGenerator(validation_split=vs)
except ValueError as e: if 'validation_split' in str(e): vs = 0.2

When it happens

Trigger: ImageDataGenerator(validation_split=0), validation_split=1.0, validation_split=-0.2, or 100 (percent instead of fraction).

Common situations: Thinking 0/1 booleans; passing a percentage (20) instead of a fraction (0.2).

Related errors

• The `factor` argument should be a number (or a list of two n
• Received: input_number={input_number}
• Received: input_number={input_number}
• Received: input_number={input_number}
• Invalid subset name: {subset}; expected "training" or "valid

AI-assisted analysis of keras-team/keras@7a34a03db6 (2026-08-25). Data as JSON: /api/errors/3a9eeabf948529b6. Report an issue: GitHub.